If all food was properly cooked, there would be no food poisoning claims. If wounds were thoroughly cleaned and disinfected the rate of severe infection would be reduced. Liability attaches to the cause of infection, not (yet) to the cause of resistance.
A new technology is needed
On the whole antibiotics do not immediately kill bacteria but instead interfere with vital repair and regeneration mechanisms. Unable to fully repair or reproduce the natural lifespan of the bacterium being a few hours leads to rapid depletion of problem colonisation and the reassertion of the dominance of the immune system. If the immune system is under-performing, any resistant bacteria and non-target organisms such as yeasts will have an open field to colonise where their competitors have been depleted.
In an ideal world, consumption of antibiotics would be accompanied by measurement of bacterial numbers, the emergence of non-target organisms and the resistance of those bacteria that remain. Consequences:
1) This would lead to cessation of consumption of antibiotics when the number of bacteria is low enough for the immune system to dominate [without irreversible damage to the patient] and
2) if the colony of resistant bacteria is growing the first antibiotic would be withdrawn and a second introduced to sufficiently damage the resistant colony for the immune system to reduce it to harmless levels.
In common practice, the duration and strength of a course of antibiotics is decided by experience or convention, its effect on bacteria is monitored via symptoms and the emergence of resistant colonies is not discovered until the re-infection doesn’t respond to the last prescription. Whilst antibiotic resistance rates were low, this non-specific practice was regarded as an acceptable trade between the simplicity of an intervention and the low frequency of complications. Antibiotic resistance rates are no longer low and back-up antibiotics are becoming less effective. It is increasingly common for combination therapies to be provided from the outset rather than wait to see if complications arise. There will come a point where non-specific interventions are no longer acceptable based on the rate of reinfection, and the generation of un-treatable colonisations. At this point a rapid and widely available test of colonisation intensity and its spectrum of antibiotic resistances combined with specific interventions would help prevent uncontrolled deterioration and the emergence of resistant colonies.
The challenge for Longitude Prize will be set to create a cheap, accurate, rapid and easy-to-use point of care test kit for bacterial infections.
The marketing material for this prize claims that life expectancy would be reduced by 20 years if antibiotics fail. This would affect life and non-life policies.
 Lifespan of asexually reproducing organisms is a tricky concept. In principle, the mother cell could live forever but in practice the probability of a fatal event in a given time window is non zero. Given enough time windows the probability of death would be high enough for a lifespan to be defined in probabilistic terms. The first bacterium could still be alive (having produced countless daughters) but probably isn’t.
 This generalization does not apply to intensive care regimes in case of septic shock.